NeurIPS 2020

Learning Multi-Agent Communication through Structured Attentive Reasoning

Meta Review

All 4 reviewers suggest that this paper is above the acceptance threshold. After reading the reviews and author response, I will recommend acceptance with a note (see bellow). Reviewers agree that the idea of filtering what information from other agents to consider within a multi-agent setup, while not new, is an important area of research, especially as the number of agents and the complexity of the environments grows. All reviewers agree that this is a sound submission, with good experimental results and comprehensive comparisons with previous work. Some concerns were raised, but the author response did a good job in addressing many of those. NOTE TO AUTHORS: Overall, all reviewers have raised very constructive and thoughtful comments, and I would like to see their points addressed in the final manuscript (basically incorporate elements of the author response in the final camera ready). In particular, the discussion about similarity to other methods, updating the result with training curves and attention metrics.